Wind power generation is typically provided by a wind “farm” of a large number (often 100 or more) wind turbine generators. Individual wind turbine generators can provide important benefits to power system operation related to mitigation of voltage flicker caused by wind gusts and mitigation of voltage deviations caused by external events.
In a wind farm setting, each wind turbine generator can experience a unique wind force. Therefore, each wind turbine generator typically includes a local controller to control the response to wind gusts and other external events. Prior art wind farm control has generally been based on one of two architectures: local control with constant power factor or reactive power and farm-level control in voltage control, or local control in constant voltage control with no farm-level control.
Local control with constant power factor and farm-level control in voltage control requires fast communications with aggressive action from the farm-level to the local level. If the farm-level control is inactive the local control can aggravate voltage flicker. With constant voltage control on each generator, steady-state operation varies significantly with small deviations in loading on the transmission grid. This causes the wind turbine generators to encounter limits in steady-state operation that prevent a response to disturbances—resulting in a loss of voltage regulation. Because reactive current is higher than necessary during this mode of operation, overall efficiency of the wind turbine generator decreases.
U.S. Pat. No. 7,224,081 describes a voltage control method and system for wind turbines wherein a reactive power regulator controls reactive power production of individual wind turbines in a wind farm by adjusting the voltage setpoint to a voltage regulator. This scheme relies on receipt of a reactive power command to each wind turbine generator. At the individual wind turbine level, a fast voltage regulator holds the wind turbine low-voltage side to a setpoint, which is adjusted by the reactive power regulator to follow the command from the wind farm control. The reactive power regulator has a first time constant that is numerically greater than a time constant of the voltage regulator. This control scheme is beneficial in that it forces all wind turbines within the wind farm to have the same reactive power output. Also, if the wind farm-level control is off, then the wind turbines all stay at a preset reactive power output even if the grid voltage varies. A downside to this scheme, however, is that the wind farm controller must act through the time constant of the reactive power regulator.